Harvard University
Cambridge, MA · private nonprofit · tier A
Your profile
80% interval on the probability, given everything we can observe about you.
Where the truth would sit depending on how your essays, recommendations and the reader's impression actually land. This is a different quantity from the interval above — not error in our estimate, but the part of the decision that is structurally invisible. It is wider, and it does not shrink with better data on our side.
Marginal effects
Percentage points of admission probability, with an 80% interval on each. Effects whose interval spans zero are reported as indistinguishable from zero, because they are.
The sample this was fit on
Coefficients estimated on 150,587 individual applicant records. Source: Document 415-1 (Arcidiacono Opening Expert Report), filed 06/15/2018, Table B.7.2: Logit estimates of Harvard's Admission decision, expanded dataset — Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, Civil Action No. 1:14-cv-14176-ADB (D. Mass.). Sample: Harvard applicants, Classes of 2014-2019 (six admission cycles, application years ~2009-2015).
| anchor fit | good — 4 published moments, worst residual 0.0159 |
| moments used | admit_rate, sat_p25, sat_p50, sat_p75 |
| cycles of admit-rate history | 10 |
| year-to-year volatility (log-odds) | 0.109 |
| transport distance from Harvard | 1.00× |
Admit rate by cycle
| year | admit rate |
|---|---|
| 2014 | 6.0% |
| 2015 | 5.6% |
| 2016 | 5.4% |
| 2017 | 5.2% |
| 2018 | 4.7% |
| 2019 | 4.6% |
| 2020 | 5.0% |
| 2021 | 4.0% |
| 2022 | 3.2% |
| 2023 | 3.5% |
No safety / target / reach label is given for this school, or any school. Those words imply a certainty this model does not have.